BiCl3-mediated opening of epoxides, a facile route to chlorohydrins or amino alcohols: One reagent, two paths

Article abstract of DOI:10.1016/j.tetlet.2005.09.088

Opening of epoxides can be an effective means by which a variety of functional groups can be incorporated. In this letter, we outline how variation of conditions, in particular, that of solvent and concentration, give rise to different products using the Lewis acid catalyst BiCl₃.

Full text of DOI:10.1016/j.tetlet.2005.09.088

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 8229–8232
BiCl3-mediated opening of epoxides, a facile route to
chlorohydrins or amino alcohols: one reagent, two paths
Adam McCluskey,a,* Sarah K. Leitch,a James Garner,a Christine E. Caden,b
Timothy A. Hill,a Luke R. Odella and Scott G. Stewarta
aChemistry, School of Environmental and Life Sciences, The University of Newcastle, Callaghan 2308, Australia
bChemistry, University of Wales Swansea, Singleton Park, Wales, SA2 8PP UK
Received 11 August 2005; revised 1 September 2005; accepted 15 September 2005
Available online 7 October 2005
Abstract—Opening of epoxides can be an effective means by which a variety of functional groups can be incorporated. In this letter,
we outline how variation of conditions, in particular, that of solvent and concentration, give rise to different products using the
Lewis acid catalyst BiCl3.
Ó 2005 Elsevier Ltd. All rights reserved.
Medicinal chemistry relies on robust, reliable reactions,
facilitating the rapid development of new chemical enti-
ties (NCEs) available for bio-evaluation. In this arena,
it is imperative that NCEs are generated rapidly, often
at the expense of synthetic elegance, perhaps more so
with the increasing levels of automation available to
the medicinal chemist. Accordingly, we are constantly
searching for, and developing, such reactions for medic-
inal applications. Recently, as part of an on-going series
of medicinal chemistry programs in our laboratory, we
developed a pharmacophore model that demanded the
synthesis of small targeted libraries of b-amino
alcohols.1,2
we favored those that occur in benign solvents, have
high atom efficiencies and those that are conducted
at room temperature and/or catalytically. It was this
latter aspect that attracted us to a recent report by
Nagaiah et al.9 in which they reported the aminolysis
ring opening of epoxides to amino alcohols catalyzed
by BiCl3. Although reports of solvent-free amine epox-
ide ring openings have been reported,3,4 many transfor-
mations have been conducted utilizing transition metals
and rare-earth triflates,10 each with their own green
chemistry issues. BiCl3 is a relatively new catalyst, which
has considerable green chemistry advantages, of low
toxicity and mild reaction conditions like many useful
bismuth based Lewis acids.9,11,12 Encouraged by these
factors, we commenced preliminary evaluations of po-
tential library assembly via a series of model reactions,
initially examining the addition of aniline, 2,4-dimethy-
laniline and 2,4-dichloroaniline to cyclohexene oxide
(Scheme 1).
There is a considerable literature precedent for the syn-
thesis of b-amino alcohols, with the conversion of epox-
ides into the corresponding trans amino alcohols being a
noteworthy example.3–5 Ring opening of epoxides is tra-
ditionally preferable to the ring opening of aziridines,6 a
function of their higher reactivity and ease of synthesis.
An additional concern in our laboratory is waste mini-
mization and the application of green chemistry princi-
ples.7 The pharmaceutical industry has a poor, but
improving record in this area with most production
processes having high E-factors.8 Consequently, in eval-
uating a potential route to our desired targeted libraries,
Our curiosity was stimulated when a product other than
the expected amino alcohol was obtained in acetonitrile.
GC–MS and NMRanalysis indicated the conversion of
cyclohexene oxide (1) to trans-2-chlorocyclohexanol (3)
OH
BiCl3
O
+
ArNH2
NHAr
Keywords: Epoxide; b-Amino alcohol; BiCl3; Green chemistry.
Corresponding author. Tel.: +61 2 4921 6486; fax: +61 2 4921
5472; e-mail: Adam.McCluskey@newcastle.edu.au
1
2
*
Scheme 1.
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.09.088

8230
A. McCluskey et al. / Tetrahedron Letters 46 (2005) 8229–8232
(Scheme 2) in the presence of all the anilines utilized in
our initial model reactions (Table 1, entries 1–3). Yields
varied between 23% and 33%, which based on the chlo-
rine stoichiometry, equates to essentially a quantitative
consumption of the available chlorine atoms in BiCl3
in this unexpected transformation.13 Increasing the reac-
tion temperature to 40 °C (Table 1, entries 4–6) had no
effect on the reaction, but it did, however, allow the
introduction of two new nucleophiles, Et3N and H2O,
which also gave rise to 3 in low yields (5% and 12%,
respectively). Conducting the reaction at 60 °C afforded
19% and 25% yields of 3 from Et3N and H2O, respec-
tively, whilst PhNH2 still provided good conversions
(based on BiCl3 content, Table 1, entries 7–9).
Having established that this is a facile conversion, we
next investigated the effect of increasing the loading of
BiCl3 to 50 mol %, now treating it as a reagent, resulted
in 94% and 72% conversion of the epoxide in the pres-
ence of PhNH2 and Et3N, respectively, to 3 (Table 1,
entries 10 and 11). These yields were now comparable
to other similar literature procedures for the production
of chlorohydrins from epoxides.14 Identification of low
yields of cyclohexane-1,2-diol (4) suggested a potential
role for trace quantities of water (in the solvent), possi-
bly coordinating with BiCl3, facilitating the release of
chlorine anion. To explore this further, CH3CN was
spiked with 200 lL of H2O (Table 1, entry 7) and the
amine removed. In this instance, 1 was completely con-
sumed with the concomitant formation of chlorohydrin
3 (69%) and a moderate, but increased, quantity of diol
4 (28%) (Scheme 2). The results tabulated below detail
that treatment of cyclohexene oxide with BiCl3/acetoni-
trile represents a mild and highly efficient route to the
corresponding chlorohydrin 3. A similar Lewis acid
mediated transformation also in the presence of acetoni-
trile/water has been reported,14e which suggests that this
solvent system may play an important part in the con-
version to the chlorohydrin species.
OH
OH
OH
Cl
BiCl3
O
+
1
3
4
Scheme 2.
Table 1. Effect of amine on yield of chlorohydrin 3, a BiCl3-mediated
epoxide opening in acetonitrile
Entry
Nucleophile
Temp
BiCl3
3 Yield
(°C)
(mol %)
(%)
1
2
3
Aniline
2,4-(CH3)2PhNH2
2,4-(Cl)2PhNH2
30
30
30
10
10
10
30
23
29
In an effort to obtain the b-amino alcohol 2 (Scheme 1),
the reaction mixture was heated at reflux (CH3CN) over-
night with PhNH2 in the presence of BiCl3 (50 mol %)
after which we noted some conversion to the amino
alcohol. Finally, the reaction with aniline was conducted
in anhydrous acetonitrile (dried over CaH2) with
20 mol % BiCl3 at room temperature (30 °C) in an effort
to use milder reaction conditions. After 36 h, we ob-
served some conversion to the amino alcohol by GC–
MS and NMRanalysis. It is interesting to note that ini-
tially 10 mol % BiCl3 was added and by GC–MS analy-
sis no reaction had occurred after 24 h. It was only when
a further 10 mol % BiCl3 was added that the reaction
took place. This does not parallel the 1–1.5 h reported
by Nagaiah and co-workers using 5 mol % catalyst.9
4
5
6
PhNH2
Et3N
H2Oa
40
40
40
10
10
10
30
5
12
7
8
9
PhNH2
Et3N
H2Oa
60
60
60
10
10
10
15
19
25b
10
11
12
PhNH2
Et3N
H2O
40
40
40
50
50
50
94
72
69c
13
14
15
PhNH2
Et3N
H2Oa
Reflux
Reflux
Reflux
10
10
10
38d
85
72e
16
17
H2Oa
H2Oa
30
0
50
50
64f
73g
Contrastingly, Ollevier et al.11 report the opening of
epoxides to afford the corresponding trans amino alco-
hols in moderate to high yields with the use of
10 mol % BiCl3 in cyclohexane. By applying this method
to cyclohexene oxide (2 M in cyclohexane), we achieved
comparable conversion to the amino alcohol within 24 h
(Table 2). In all cases, a minor product was observed
a 200 lL H2O added.
b 9% diol 4.
c 28% diol 4.
d 0.4% diol 4 + 18% amino alcohol.
e 16% diol 4.
f 22% diol 4.
g 9% diol 4.
Table 2. Conversion to amino alcohol 2 in various solvents
Entry
Reactant
Solvent
Temp (°C)
Concentrationa
BiCl3 (mol %)
Product ratios 2:3:4
1
2
3
4
5
6
Aniline
Aniline
Aniline
Aniline
Aniline
H2O
CH3CN
CH3CNb
CH3CN
DCM
Cyclohexane
CH3CN
Reflux
0.1 M
0.1 M
2 M
2 M
2 M
10
10
10
10
10
10
0:4.4:1
1:2:0
1:1:0
1.7:1:0
5.4:1:0
0:2:1
rt
rt
rt
rt
rt
2 M
a Concentration with respect to starting epoxide.
b Anhydrous CH3CN, dried over CaH2, reaction left for 36 h.

A. McCluskey et al. / Tetrahedron Letters 46 (2005) 8229–8232
8231
that being the chlorohydrin. However, this was signifi-
cantly lower when compared to using acetonitrile as a
solvent.
perature. Interestingly, at this temperature the more ste-
rically hindered 1-adamantylamine had one of the
highest turnovers (Table 4, entry 5), while in each reac-
tion conducted a large amount of starting material still
remained. Increasing the amine stoichiometry and/or
catalyst loading, in an effort to improve the yield of
these reactions, failed to produce large amounts of
the b-amino alcohol. However, increasing the tempera-
ture to 40 °C improved the turnover, especially in the
case of propylamine, improving the yield from 24% to
91%.
Epoxide ring opening involving a series differently
substituted anilines was also examined (Table 3).15 As
previously mentioned, substrates, cyclohexene oxide 1
(Scheme 1) and styrene oxide 5 (Scheme 3) were tested.
Good conversion to the amino alcohols (2 and 6) can be
obtained, even when extremely electron-deficient ani-
lines are used (i.e., 2,4-dinitroaniline, Table 3, entry 5).
Slightly lower yields were observed when a more steri-
cally hindered aniline, 2,6-dimethylaniline, was used
(Table 3, entries 3 and 9). Similarly, a series non-aro-
matic amines were also added to styrene oxide (Table
4). Unfortunately, turnover to the b-amino alcohol
was slow when reactions were carried out at room tem-
Summary: We have demonstrated that bismuth(III)
chloride in acetonitrile can be used as a reagent to facili-
tate the formation of chlorohydrin 3 in excellent yield
from the corresponding epoxide 1, even in the presence
of aniline. Conversely, a change in solvent to cyclohex-
ane gives the expected b-amino alcohol 2 or 6, BiCl3
in this case acting as chelating Lewis acid. These results
are currently being incorporated into our program
of developing scaffolds for drug design on which a full
paper is forthcoming.
Table 3. Reaction of styrene and cyclohexene oxide with 10% BiCl3
and various anilines: synthesis of amino alcohols 2 and 6
Entry
Ar-NH2
Yield (%)
Cyclohexene oxide
1
2
3
4
5
6
Ar=Ph
o-Et-Ph
2,6-(CH3)2-Ph
2-CH3-6-NO2-Ph
2,4-(NO2)2-Ph
2,4-(Cl)2-Ph
89
92
73
90
75
91
Acknowledgments
We thank the NH&MRC, for funding toward this
project.
References and notes
Styrene oxide
7
8
9
10
11
Ar=Ph
o-Et-Ph
2,6-(CH3)2-Ph
3-Cl-Ph
2,4-(Cl)2-Ph
78
76
62
82
64
1. Keller, P. A.; Bowman, M.; Dang, K.-H.; Leach, S. P.;
Garner, J.; Smith, R.; McCluskey, A. J. Med. Chem. 1999,
42, 2351.
2. McCluskey, A.; Keller, P. A.; Morgan, J.; Garner, J. Org.
Biomol. Chem. 2003, 1, 3353.
3. Kamal, A.; Ramu, R.; Azhar, M. A.; Khanna, G. B. R.
Tetrahedron Lett. 2005, 46, 2675.
4. Zhao, P.-Q.; Xu, L.-W.; Xia, C.-G. Synlett 2004, 846.
Ar
´
5. (a) Carree, F.; Gil, R.; Collin, J. Tetrahedron Lett. 2004,
Cl
HN
Ph
BiCl3 (10 mol%)
cHexane
O
45, 7749; (b) Sabitha, G.; Reddy, G. S.; Reddy, K. B.;
Yadev, J. S. Synthesis 2003, 2298; (c) Yadev, J. S.; Reddy,
B. V. S.; Basak, A. K.; Narasaiah, A. V. Tetrahedron Lett.
Ar NH2
+
+
Ph
Ph
OH
OH
´
2003, 44, 1047; (d) Pachon, L. D.; Gamez, P.; van Brussel,
6
5
7
J. J. M.; Reedijk, J. Tetrahedron Lett. 2003, 44, 6025; (e)
Rampalli, S.; Chaudhari, S. S.; Akamanchi, K. G.
Synthesis 2000, 78; (f) Das, U.; Crousse, B.; Kesavan,
Scheme 3.
´
´
V.; Bonnet-Delpon, D.; Begue, J.-P. J. Org. Chem. 2000,
65, 6749; (g) Chng, B. L.; Ganesan, A. Bioorg. Med. Chem.
Lett. 1997, 7, 1511.
Table 4. Reaction of styrene oxide with various amines and 10% BiCl3:
synthesis of amino alcoholsa
´
6. (a) Concellon, J. M.; Riego, E. J. Org. Chem. 2003, 68,
Entry
Amine
Temperature
(°C)
% Conversionb
(SM)c
6407; (b) Tamamura, H.; Yamashita, M.; Nakajima, Y.;
Sakano, K.; Otaka, A.; Ohno, H.; Ibuka, T.; Fujii, N. J.
Chem. Soc., Perkin Trans. 1 1999, 2983.
1
2
3
4
5
6
7
8
n-Propylamine
n-Propylamine
n-Butylamine
rt
40
rt
40
rt
40
rt
20 (74)
91 (4)
´
7. (a) Wardencki, W.; Curyło, J.; Namiesnik, J. Pol. J.
Environ. Stud. 2005, 4, 389; (b) Andraos, J. J. Org. Process
Res. Dev. 2005, 9, 404; (c) Goodwin, T. E. J. Chem. Edu.
2004, 81, 1187.
11 (57)
30 (11)
34 (50)
35 (55)
2 (95)
n-Butylamine
1-Adamantylamine
1-Adamantylamine
Morpholine
8. Sheldon, R. A. Comptes Rendus de l’Academie des
Sciences, Serie IIc: Chimie 2000, 3, 541.
9. Swamy, N. R.; Kondaji, G.; Nagaiah, K. Synth. Commun.
2002, 32, 2307.
10. Sekar, G.; Singh, V. K. J. Org. Chem. 1999, 64, 287.
11. (a) Ollevier, T.; Lavie-Compin, G. Tetrahedron Lett. 2004,
45, 49; (b) Ollevier, T.; Lavie-Compin, G. Tetrahedron
Lett. 2002, 43, 7891.
Morpholine
40
14 (18)
a Reaction carried out in cyclohexane.
b % Conversion to the two amino alcohol regioisomers as determined
by GC–MS.
c Starting material remaining.

8232
A. McCluskey et al. / Tetrahedron Letters 46 (2005) 8229–8232
12. (a) Anzalone, P. W.; Baru, A. R.; Danielson, E. M.;
Hayes, P. D.; Nguyen, M. P.; Panico, A. F.; Smith, R. C.;
Mohan, R. S. J. Org. Chem. 2005, 70, 2091; (b) Gaspard-
Iloughmane, H.; Le Roux, C. Eur. J. Org. Chem. 2004,
2517; (c) Antoniotti, S.; Dunach, E. C. R. Chimie 2004,
679; (d) Khosropour, A. R.; Khodaei, M. M.; Ghozati, K.
Russ. J. Org. Chem. 2003, 40, 1332; (e) Anderson, A. M.;
Blazek, J. M.; Garg, P.; Payne, B. J.; Mohan, R. S.
Tetrahedron Lett. 2000, 41, 1527.
14. (a) Ranu, B. C.; Banerjee, S. J. Org. Chem. 2005, 70, 4517;
(b) Xu, L.-W.; Li, L.; Xia, C.-G.; Zhao, P.-Q. Tetra-
hedron Lett. 2004, 45, 2435; (c) Smitha, G.; Reddy, C. S.
J. Chem. Res. 2004, 300; (d) Sartillo-Piscil, F.; Quintero,
´
L.; Villegas, C.; Santacruz-Juarez, E.; de Parrodi, C. A.
Tetrahedron Lett. 2002, 43, 15; (e) Sabitha, G.; Babu, R.
S.; Rajkumar, M.; Reddy, C. S.; Yadev, J. S. Tetrahedron
Lett. 2001, 42, 3955.
1
15. NMRdata for aniline 6 (Table 3, Entry 7)5f: H NMR:
13. (a) Denmark, S. E.; Wynn, T.; Jellerichs, B. G. Angew.
Chem., Int. Ed. 2001, 40, 2256; (b) NMRdata for chloro-
hydrin 3: 1H NMR: (300 MHz) d (ppm) 1.24–1.32 (m, 3H,
CH2), 1.51–1.70 (m, 1H, CH2), 1.72–1.76 (m, 2H, CH2),
2.09–2.10 (m, 1H, CH2), 2.11–2.23 (m, 1H, CH2), 2.49 (br
s, 1H, OH), 3.42–3.52 (m, 1H, CH), 3.66–3.74 (m, 1H,
CH). 13C NMR(75.5 MHz) d (ppm) 24.0 (CH2), 25.7
(CH2), 33.1 (CH2), 35.1 (CH2), 67.5 (CH), 75.4 (CH).
(300 MHz) d (ppm) 3.81 (dd, J = 11.3 and 7.62 Hz, 1H,
CH2), 3.93 (dd, J = 11.3 and 7.2 Hz, 1H, CH2), 4.52 (dd,
J = 7.3 and 4.1 Hz, 1H, CH) 6.69 (d, J = 8.3 Hz, 1H, CH-
Ar), 6.76 (m, 1H, CH-Ar), 7.10–7.40 (m, 8H, CH-Ar). 13
C
NMR(75.5 MHz) d (ppm) 60.7 (CH), 66.1 (CH2), 114.6
(CH), 118.6 (CH), 126.5 (CH), 126.6 (CH), 127.2
(CH), 128.2 (CH), 128.6 (CH), 128.7 (CH), 138.6 (C),
145.2 (C).